a8f20420a
Denys Vlasenko
2025-08-14
Subject
hush: shrink syntax error handling
Binary
No binary (build failed)
Architecture & Compiler Flags
—
Build Log
(log yok)
Commit Diff GitHub →
shell/hush.c+52 -28
| @@ -738,7 +738,7 @@ struct parse_context { | |
| #if !BB_MMU | |
| o_string as_string; | |
| #endif | |
| - | smallint is_assignment; /* 0:maybe, 1:yes, 2:no, 3:keyword */ |
| + | smallint is_assignment; |
| #if HAS_KEYWORDS | |
| smallint ctx_res_w; | |
| /* bitmask of FLAG_xxx, for figuring out valid reserved words */ | |
| @@ -4051,10 +4051,10 @@ static int done_pipe(struct parse_context *ctx, pipe_style type) | |
| ctx->ctx_res_w = RES_NONE; | |
| #endif | |
| #if ENABLE_HUSH_CASE | |
| - | if (ctx->ctx_res_w == RES_MATCH) |
| - | ctx->ctx_res_w = RES_CASE_BODY; |
| if (ctx->ctx_res_w == RES_CASE) | |
| ctx->ctx_res_w = RES_CASE_IN; | |
| + | if (ctx->ctx_res_w == RES_MATCH) |
| + | ctx->ctx_res_w = RES_CASE_BODY; |
| #endif | |
| /* Create the memory for command, roughly: | |
| * ctx->pipe->cmds = new struct command; | |
| @@ -4415,6 +4415,7 @@ static int done_word(struct parse_context *ctx) | |
| if (ctx->is_assignment != DEFINITELY_ASSIGNMENT) { | |
| ctx->is_assignment = NOT_ASSIGNMENT; | |
| } else { | |
| + | /* This was an assignment word, next one maybe will be too */ |
| command->assignment_cnt++; | |
| debug_printf_parse("++assignment_cnt=%d\n", command->assignment_cnt); | |
| ctx->is_assignment = MAYBE_ASSIGNMENT; | |
| @@ -4426,7 +4427,7 @@ static int done_word(struct parse_context *ctx) | |
| #if ENABLE_HUSH_LOOPS | |
| if (ctx->ctx_res_w == RES_FOR) { | |
| if (ctx->word.has_quoted_part | |
| - | || endofname(command->argv[0])[0] != '\0' |
| + | || endofname(ctx->word.data)[0] != '\0' |
| ) { | |
| /* bash says just "not a valid identifier" */ | |
| syntax_error("bad for loop variable"); | |
| @@ -4441,9 +4442,12 @@ static int done_word(struct parse_context *ctx) | |
| #endif | |
| #if ENABLE_HUSH_CASE | |
| /* Force CASE to have just one word */ | |
| - | if (ctx->ctx_res_w == RES_CASE) { |
| + | if (ctx->ctx_res_w == RES_CASE) |
| done_pipe(ctx, PIPE_SEQ); | |
| - | } |
| + | //TODO syntax check? |
| + | // if (ctx->ctx_res_w == RES_MATCH) |
| + | // eat all following spaces and tabs (but not newlines), |
| + | // check that next char is | or ) |
| #endif | |
| ret: | |
| @@ -5800,7 +5804,7 @@ static struct pipe *parse_stream(char **pstring, | |
| */ | |
| if (ch == '}') { | |
| if (!IS_NULL_WORD(ctx.word)) { | |
| - | /* word} or ""} */ |
| + | /* WORD} or ""} */ |
| goto ordinary_char; | |
| } | |
| if (!IS_NULL_CMD(ctx.command)) { /* CMD } */ | |
| @@ -5924,8 +5928,8 @@ static struct pipe *parse_stream(char **pstring, | |
| /* note: we do not add it to &ctx.as_string */ | |
| /* TODO: in bash: | |
| * comment inside $() goes to the next \n, even inside quoted string (!): | |
| - | * cmd "$(cmd2 #comment)" - syntax error |
| - | * cmd "`cmd2 #comment`" - ok |
| + | * CMD "$(CMD2 #COMMENT)" - syntax error |
| + | * CMD "`CMD2 #COMMENT`" - ok |
| * We accept both (comment ends where command subst ends, in both cases). | |
| */ | |
| while (1) { | |
| @@ -5945,7 +5949,7 @@ static struct pipe *parse_stream(char **pstring, | |
| rbrace_skips_end_trigger: | |
| if (ctx.is_assignment == MAYBE_ASSIGNMENT | |
| - | /* check that we are not in word in "a=1 2>word b=1": */ |
| + | /* check that we are not in WORD in "a=1 2>WORD b=1": */ |
| && !ctx.pending_redirect | |
| ) { | |
| /* ch is a special char and thus this word | |
| @@ -6019,7 +6023,7 @@ static struct pipe *parse_stream(char **pstring, | |
| ctx.ctx_res_w = RES_MATCH; /* "we are in PATTERN)" */ | |
| } | |
| #endif | |
| - | new_cmd: |
| + | finished_cmd: |
| /* We just finished a cmd. New one may start | |
| * with an assignment */ | |
| ctx.is_assignment = MAYBE_ASSIGNMENT; | |
| @@ -6049,14 +6053,24 @@ static struct pipe *parse_stream(char **pstring, | |
| goto parse_error_exitcode1; | |
| } | |
| } | |
| - | goto new_cmd; |
| + | goto finished_cmd; |
| case '|': | |
| - | if (done_word(&ctx)) |
| - | goto parse_error_exitcode1; |
| #if ENABLE_HUSH_CASE | |
| - | if (ctx.ctx_res_w == RES_MATCH) |
| - | break; /* we are in case's "WORD | WORD)" */ |
| + | if (ctx.ctx_res_w == RES_MATCH) { |
| + | // if (IS_NULL_WORD(ctx.word)) { |
| + | // /* Testcase: sh -c 'case w in |w) nice; esac' */ |
| + | // /* Testcase: sh -c 'case w in v||w) nice; esac' */ |
| + | //also rejects valid: sh -c 'case w in v |w) nice; esac' |
| + | // syntax_error_unexpected_ch(ch); |
| + | // goto parse_error_exitcode1; |
| + | // } |
| + | if (done_word(&ctx)) |
| + | goto parse_error_exitcode1; |
| + | continue; /* get next char */ |
| + | } |
| #endif | |
| + | if (done_word(&ctx)) |
| + | goto parse_error_exitcode1; |
| if (next == '|') { /* || */ | |
| ch = i_getch(input); | |
| nommu_addchr(&ctx.as_string, ch); | |
| @@ -6079,45 +6093,55 @@ static struct pipe *parse_stream(char **pstring, | |
| } | |
| done_command(&ctx); | |
| } | |
| - | goto new_cmd; |
| + | goto finished_cmd; |
| case '(': | |
| #if ENABLE_HUSH_CASE | |
| - | /* "case... in [(]PATTERN)..." - skip '(' */ |
| + | /* "case... in (PATTERNS)..."? */ |
| if (ctx.ctx_res_w == RES_MATCH | |
| - | && ctx.command->argv == NULL /* not (PATTERN|(... */ |
| - | && IS_NULL_WORD(ctx.word) /* not PATTERN(... or ""(... */ |
| + | && ctx.command->argv == NULL /* not WORD (... */ |
| + | && IS_NULL_WORD(ctx.word) /* not WORD(... or ""(... */ |
| ) { | |
| - | continue; /* get next char */ |
| + | continue; /* skip '(', get next char */ |
| } | |
| #endif | |
| /* fall through */ | |
| case '{': { | |
| + | /* Try to parse as { CMDS; } or (CMDS) */ |
| int n = parse_group(&ctx, input, ch); | |
| if (n < 0) | |
| goto parse_error_exitcode1; | |
| debug_printf_heredoc("parse_group done, needs heredocs:%d\n", n); | |
| heredoc_cnt += n; | |
| - | goto new_cmd; |
| + | goto finished_cmd; |
| } | |
| case ')': | |
| #if ENABLE_HUSH_CASE | |
| if (ctx.ctx_res_w == RES_MATCH) { | |
| + | // if (IS_NULL_WORD(ctx.word)) { |
| + | // /* Testcase: sh -c 'case w in w|) nice; esac' */ |
| + | //also rejects valid: sh -c 'case w in w ) nice; esac' |
| + | // syntax_error_unexpected_ch(ch); |
| + | // goto parse_error_exitcode1; |
| + | // } |
| if (done_word(&ctx)) | |
| goto parse_error_exitcode1; | |
| + | //FIXME: accepts "W1 W2) CMD;;" as if it was "W1|W2) CMD;;", should not allow this syntax at all |
| done_pipe(&ctx, PIPE_SEQ); | |
| - | goto new_cmd; |
| + | goto finished_cmd; |
| } | |
| #endif | |
| - | case '}': |
| + | /* case ')' and !RES_MATCH: falls through to the error: */ |
| + | /* Testcase for bad ')' ? */ |
| + | /* case '}': */ |
| /* Proper use of this character is caught by end_trigger: | |
| * if we see {, we call parse_group(..., end_trigger='}') | |
| - | * and it will match } earlier (not here). */ |
| + | * and it will match } earlier (not here). |
| + | */ |
| + | /* Testcase for bad '}' ? */ |
| + | default: /* all other chars should not ever reach this... */ |
| G.last_exitcode = 2; | |
| syntax_error_unexpected_ch(ch); | |
| goto parse_error; | |
| - | default: |
| - | if (HUSH_DEBUG) |
| - | bb_error_msg_and_die("BUG: unexpected %c", ch); |
| } | |
| } /* while (1) */ | |
| eof: | |